Actions

Ontolog Forum

Revision as of 20:11, 20 September 2017 by imported>KennethBaclawski
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Ontology Summit 2018 Second Research Session

Followup Discussion

[13:32] DouglasRMiles: Organizing contexts thru genlMts is interesting

[13:33] janet singer: It would be good to have issues on this topic continued and then have the new bag of worms get full attention separately

[13:35] David Whitten: Yes Doug, if you consider a context as a computed set of propositions. Another issue is the salient assumptions on a microtheory.

[13:36] David Whitten: The salient assumptions are assumed to be "AND"ed with each individual proposition in the Microtheory.

[13:38] DouglasRMiles: Right Davind, I been thinking genllVocabularyMt and genlMt will havbe to be used

[13:38] DouglasRMiles: as in what actually gets lifted between Mts

[13:40] DouglasRMiles: oh genlVocabularyMt is how we import a vocab

[13:40] DouglasRMiles: but what i meant is that you inherit only vocab a Mt is importoring

[13:41] DouglasRMiles: (in summary what we filter in the lifting of Mts cycorp had it right in that there is probably about 12 differnt ways that lifting has to be specialized)

[13:42] DouglasRMiles: I suppose it could be claimed that ResearchCyc impls IKL

[13:43] David Whitten: yeah. probably. Did Cycorp actually include that list of ways to lift in a public paper? I know R.V.Guha defined one in his doctorate thesis.

[13:43] John Sowa: Re cyc: In 2006, Doug Lenat agreed that IKL was sufficient to express CycL.

[13:44] David Whitten: I don't know if "that" is really a single IMPLEMENTATION relation. There may be separate strategies for different usages of it. It looks to have a complex definition for how it quotes a proposition.

[13:44] John Sowa: But Lenat also said that Common Logic without the 'that' operator was not sufficient to represent Cycl.

[13:45] David Whitten: Thanks John. I'm glad Doug Lenat said that. It raises hope that IKL can be implemented by mere humans.

[13:45] John Sowa: If you look at the IKRIS final report, you'll see the results of using IKL as the interchange language among several systems including Cyc.

[13:45] David Whitten: CycL has quotedIsa and other quote mechanisms.

[13:46] David Whitten: It's not obvious to me that exceptWhen and other exception syntax can be implemented without "that".

[13:47] John Sowa: A good property of IKL: you get all the power of CycL, modal logics, context logics, situation theory, etc., with just one operator added to Common Logic: 'that'.

[13:47] DouglasRMiles: Yeah Actually 'that' i suppose distiguishes a CycLAssertion from a CycSentence

[13:47] David Whitten: Use of a surface syntax does not always show what is necessary to implement the "deep syntax" which is part of the surface syntax's meaning.

[13:48] DouglasRMiles: sorry: CycLSentence-Assertable from a CycAssertion

[13:48] David Whitten: John, now that you are on chat, what is ICL ?

[13:48] DouglasRMiles: btw 28 specs of genlMt http://logicmoo.org:3602/cgi-bin/cg?cb-lq&c6137&2&c18643

[13:49] David Whitten: I agree re DouglasRMiles: distinguishing between CycLSentence-Assertable and CycAssertion

[13:51] John Sowa: David, where did you find the abbreviation ICL? That is not one I used.

[13:52] DouglasRMiles: When chatting with some friends of mine whom I think have a good understanding of mathmatical logic.. IKL is very hard to introduce

[13:52] DouglasRMiles: Since it is a knowledge representation language as much as it could be called a logic

[13:53] DouglasRMiles: well trying to call it a logic is dangerous anyways :P

[13:53] David Whitten: ICL was in a URL from the IKRIS project. I'll find the URL and give it too you.

[13:54] DouglasRMiles: but still sometimes i wish i could give IKL a name "This is a HOL language"

[13:54] David Whitten: John, is your =P operator definable in Common Logic or does it require IKL ?

[13:55] John Sowa: Re =P operator: That was Pat's term. I would just say that the operator '=' is equality.

[13:56] DouglasRMiles: Pat Hayse defines it as a "unsorted FOL" moment i say first order i think people are beginning to distinguish "sorts"

[13:56] David Whitten: ah. so =P is just = in an IKL system. Interesting.

[13:57] John Sowa: When you write {= P (that (on cat mat)) ), I would just say that the equality states that P is

[13:58] John Sowa: P is the name of the proposition that is named by (that (on cat mat))

[13:59] John Sowa: I deleted many of Pat's slides that would take too long to explain. I should have deleted the one that called "=P" an operator.

[13:59] DouglasRMiles: oh iot wasnt just a missing space

[14:00] John Sowa: I would just say that the operator is the plain old equality '='

[14:00] DouglasRMiles: (and (= P (that (on cat mat)) ) (ist CatTheories P))

[14:01] DouglasRMiles should have quantified to show which are vars but everyone gets it

[14:02] John Sowa: In any case, I have to sign off now.

[14:02] DouglasRMiles: "getting it" as /me was given an example why someone might do that

[14:02] DouglasRMiles: thank you John

[14:03] DouglasRMiles: glad you do this for us!

[14:04] DouglasRMiles: scary thought about what the world would be without

[14:04] David Whitten: John, I'll try to find that URL by next week. I can't find it in the flurry today.

[14:05] DouglasRMiles: ive seen the words ICL in places.. I sorta felt like MELD has undergone so many names .. but it all would refer to the same KR language eventualyl

[14:05] David Whitten: Yeah.the world is nicer when people share definitions with you, so you don't have to start from scratch. I've been there before.

[14:07] David Whitten: re marking variables, I like the CG practice of distinguishing the "free" variable and the "bound" variable using different prefixes.

[14:07] DouglasRMiles: what i mean is I saw "ICL" used at least once on Pat or Johns site.. but it was I assumed one name used in a project like for about 3 months in 2002

[14:08] DouglasRMiles: And by just one of the many IKRIS projects people whom forgot to rename their proparity lanauge that have been intially names ICL in 2002 to IKL for 20xx of the IKRIS project

[14:09] DouglasRMiles: Like me forgetting to rename cycL in my docs to IKL (hehe)

[14:10] DouglasRMiles: I really hope IKL can be ISO-ifed or W3C-ified

[14:11] DouglasRMiles: not to replay a usual rant.. But in RoboKind software I used RDF and the funder understood that

[14:13] DouglasRMiles: I want to be able to say I used IKL and have the weight of Tim Berners Lee or someone whom has endorsed it to.. Thogh Pat Hayse and John Sowa of course says a lot

[14:14] DouglasRMiles: but a committee endorsment is more what i meant with TimBL

[14:15] DouglasRMiles: I am going to try to start leaving this chat open all the time David

[14:15] DouglasRMiles: I hope you and others do to

[14:16] DouglasRMiles makes lunch.. is AFKish

[14:17] DouglasRMiles: It be interesting if people say things about themselves so doent have to google them after lunch